[Insight-developers] ITK not redistributable: copyright statement
incomplete
Luis Ibanez
luis.ibanez at kitware.com
Sat Sep 23 20:02:36 EDT 2006
Hi Steve,
Thanks for pointing this out.
This is very timely to our discussion on how to accept new code
into the Insight Toolkit. This Copyright issues are of grave
importance. They are indeed one of few factor that threaten
the evolution of Open Source software. Most the of the copyright
issues that you listed are the result of a lax policy on accepting
new code into ITK.
-----
There seem to be multiple problems, with the copyright notices,
so we may have to arrange for a combination of solutions.
Let me suggest here some options for
each one of the cases that you listed:
0) VXL and VTK: I would suggest that we add more visible credits
to both VTK and VXL. That may imply to retouch
the copyright headers of *all* ITK files.
1) DICOMParser: It is a deprecated code. We don't need it now
that GDCM provides much better DICOM support.
These code should never have been added to ITK
without a copyright transferm to ISC. That being
said, we thank Matt Turek for allowing us to
distribute this code with ITK.
I vote to remove it before the release of ITK 3.0.
2) AnalyzeDhb: Mayo Clinic Copyrights.
I suggest we ask for help from Hans Johnson and
Kent Williams to see if the Mayo Clinic can grant
a BSD-like license to the ISC in order to distribute
this code.
If no copyright transfer is possible we should remove
this code from ITK, and find another way of imple
menting Analyze readers.
3) itkPixelData: GE Copyrights.
I suggest we ask for help to Bill Lorensen and
Jim Miller to see if a copyright transfer is possible
in this case, or whether GE can grant a BSD-like
license to the ISC for distributing this code.
Note that this header is only used by the DICOMParser
in(1) to be deprecated, and by the GE5ImageIO.
If no copyright transfer is possible we should remove
this code from ITK. This code is of secondary impor-
tance and will affect a minority of users.
4) itkOctreeNode.h/cxx: University of Iowa, Vincent Magnotta
I suggest we ask for help from Vincent Magnotta, for
transfering the copyright to the ISC, or to provide
a BSD-license to the ISC for distributing this code.
If not copyright transfer, or license is possible,
this code should be removed from ITK before ITK 3.0.
5) itkThresholdMaximumConnectedComponentsImageFilter: Ken Urish.
This one is my mistake. I uploaded this code from
Ken's contribution to the Insight Journal, and skipped
the formality of asking for a copyright transfer.
I will follow with Ken if he is willing to transfer
the copyright to the ISC, or provide a BSD-Like license
to the ISC for distributing the code.
If not copyright transfer, or license is possible,
this code should be removed from ITK before ITK 3.0.
6) Utilities/vxl/v3p/netlib: Netlib and IMSL stuff, no license
This one is surprisingly tricky. The web site:
www.netlib.org
has very sparse information about copyrights and
licensing. There is however a LAPACK Debian
package already, so we probably only need to
add credits to netlib here...
7) triangle.h/c: Copyright Jr Shewchuk,
The license excludes commercial use.
This is in complete contradiction to the BSD-Like
license of ITK.
The program triangle.c is actually a 2D Mesh generator
for Delaunay Triangulations. There is no reason for
having such functionality in ITK.
I just removed the files from CVS.
Please let us know if you find these options to be reasonable,
and if I missed any of the issues that you pointed out.
Thanks,
Luis
================================================
> Triangle stuff copyright JR Shewchuk, only free for noncommercial and
> must include full modified source
-----------------------
Steve Robbins wrote:
> Hello,
>
> Gavin Baker and I have been working for some time to make Debian
> packages of ITK. The technical issues have been cleared up and the
> remaining hurdle is the copyright statement.
>
> The ITK copyright statement (http://www.itk.org/HTML/Copyright.htm) is
> incomplete. It leads one to believe that the entire ITK source tree is
> copyrighted by the consortium, which is absolutely untrue.
>
> The "Copyright.txt" file of the CVS sources is better. This text
> acknowledges contributions from VTK and from VXL:
>
> The Insight Segmentation and Registration Toolkit (ITK) was initially
> developed under contract to the National Library of Medicine at the
> National Institutes of Health. ITK is partially derived from VTK and VXL,
> hence some code is copyrighted accordingly (see VTKCopyright.txt and
> VXLCopyright.txt).
>
> However, even this is not the whole story. There are a number of other
> copyright/license statements not mentioned, such as those of
> directories Utilities/kwsys, Utilities/DICOMParser, and Utilities/png.
> Bits of the core ITK fall under all kinds of other copyrights, too.
> Examples include: itk_hashtable.h, itk_hash_map.h, itk_hash_set.h,
> itkMersenneTwisterRandomVariateGenerator.h, and vcl_bvector.h
>
> Fortunately, these licenses are all compatible with ITK, at least to my
> un-lawyerly eye, so these omissions are easy to fix.
>
> The real problem is that there are a number that are NOT compatible
> with ITK license in that no license to redistribute is granted:
>
> Code/IO/itkPixelData.h:
> Copyright (C) 1987, 1988, 1989 The General Electric Company
>
> Code/IO/itkAnalyzeDbh.h
> Has parts copyright under copyright by the Mayo Foundataion
>
> Code/Common/itkOctreeNode.cxx
> parts * Copyright (C) 2002 Iowa Mental Health Clinical Research Center
>
> Code/Common/itkOctreeNode.h
> Copyright (C) 1998 Nancy C. Andreasen and Vincent A. Magnotta
>
> itkThresholdMaximumConnectedComponentsImageFilter.h
> itkThresholdMaximumConnectedComponentsImageFilter.txx
> Copyright (c) Ken Urish 2005. All rights reserved.
>
> Utilities/vxl/v3p/netlib
> Netlib and IMSL stuff, no license
> Triangle stuff copyright JR Shewchuk, only free for noncommercial and
> must include full modified source.
>
>
> I stopped looking at this point. :-(
>
> What is the procedure to fix inconsistent licensing like this?
>
> Thanks,
> -Steve
>
> _______________________________________________
> Insight-developers mailing list
> Insight-developers at itk.org
> http://www.itk.org/mailman/listinfo/insight-developers
>
>
More information about the Insight-developers
mailing list